Closed aPovidlo closed 9 months ago
Hello @aPovidlo! Thanks for updating this PR. We checked the lines you've touched for PEP 8 issues, and found:
There are currently no PEP 8 issues detected in this Pull Request. Cheers! :beers:
Merging #1155 (9865e34) into master (a5a9b17) will decrease coverage by
0.09%
. The diff coverage is72.72%
.
@@ Coverage Diff @@
## master #1155 +/- ##
==========================================
- Coverage 79.54% 79.45% -0.09%
==========================================
Files 142 144 +2
Lines 9805 9954 +149
==========================================
+ Hits 7799 7909 +110
- Misses 2006 2045 +39
Files | Coverage Δ | |
---|---|---|
fedot/api/api_utils/presets.py | 87.50% <100.00%> (ø) |
|
fedot/core/data/data_preprocessing.py | 93.50% <100.00%> (+0.26%) |
:arrow_up: |
fedot/core/data/data_split.py | 94.73% <ø> (ø) |
|
...ore/operations/evaluation/evaluation_interfaces.py | 86.48% <ø> (-0.13%) |
:arrow_down: |
...mentations/data_operations/categorical_encoders.py | 98.07% <100.00%> (+0.07%) |
:arrow_up: |
fedot/preprocessing/preprocessing.py | 97.00% <100.00%> (+0.01%) |
:arrow_up: |
fedot/core/data/visualisation.py | 15.49% <0.00%> (ø) |
|
fedot/core/operations/evaluation/boostings.py | 76.31% <76.31%> (ø) |
|
fedot/core/data/data.py | 70.46% <65.85%> (-0.82%) |
:arrow_down: |
...mplementations/models/boostings_implementations.py | 72.30% <72.30%> (ø) |
First step in global boosting methods refactoring. Taking out Catboost model from SkLearnEvaluationStrategy into separate Boosting class strategy.
New features for CatBoost:
Updates in data, preprocessing and data_split:
Results: OLD VERSION - {'Scoring': 0.5548, 'Flight': 0.5613} NEW VERSION - {'Scoring': 0.5821, 'Flight': 0.5806}